Digital dental implant bridge and false tooth and false tooth automatic adaptation and production method and system based on CBCT (cone beam computed tomography) and oral scanning data

Abstract

The invention discloses a digital dental implant bridge and false tooth and false tooth automatic adaptation and production system based on CBCT (cone beam computed tomography) and oral scanning data and a control method thereof, which are used for realizing automatic design, adaptation and production of a dental implant bridge and false teeth and false teeth and fusing the CBCT data and the oral scanning data to generate a three-dimensional oral cavity digital model. An automatic modeling technology based on deep learning is used for analyzing the occlusion relation, the adjacent tooth form and the implant angle of a patient, and an implant tooth bridge and false teeth are automatically designed to obtain a three-dimensional model of the prosthesis. A computer vision technology or a three-dimensional scanning technology is adopted to detect the consistency of the size and shape of a produced finished product and the three-dimensional model; and digital try-on is performed through an augmented reality (AR) technology, so that rapid adjustment of various parameters of the product is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of dental implant data processing, and particularly relates to a method and system for automatically adapting and producing digital dental implant bridges and denture prostheses based on CBCT and intraoral scan data. Background Art

[0002] A denture is an artificial tooth that replaces a missing natural tooth and is a prosthesis made after partial or complete loss of the upper and lower teeth. After a tooth is missing, a thing similar to a bridge installed on both sides of the missing tooth is called a dental bridge. A dental bridge can repair the gap formed by one or more missing teeth in the dental arch. A dental bridge consists of the crowns of two or more teeth on both sides of the missing tooth and the artificial tooth located between them. The teeth that play a fixing role are called " abutment teeth", and the artificial tooth is called " pontic". A dental bridge can be made of gold, alloy, porcelain or a combination of these materials, and the required supporting force is provided by natural teeth or implants.

[0003] The existing production process of dental implant bridges and denture prostheses usually relies on the adaptation design after oral model printing, which is time-consuming and has complex procedures. Therefore, the present invention proposes a digital full process, which combines CBCT and intraoral scan data, eliminates the traditional model making link, and realizes a personalized, precise and automated implant restoration method.

[0004] Among them, CBCT (Cone beam CT) is a cone beam projection computerized tomography imaging device. Its principle is that the X-ray generator makes a circular DR (digital projection) around the projection object with a relatively low radiation dose (usually about 10 milliamperes of tube current), and then the data obtained by intersecting the digital projections around the projection object multiple times (usually 180 to 360 times) is reconstructed in the computer to obtain a three-dimensional image.

[0005] Oral scanning is to obtain the overall situation of teeth by using an intraoral scanner after understanding the tooth position through oral examination and radiography during orthodontics. Oral scanning has the functions of comfortable impression taking and real-time effect display. Usually, the oral scanning result is stored as an STL file, and the STL file is used to describe the surface geometry of a three-dimensional object. Summary of the Invention

[0006] In view of this, it is necessary to provide a convenient, efficient and intelligent method and system for automatically adapting and producing digital dental implant bridges and denture prostheses based on CBCT and intraoral scan data.

[0007] A method for automatically adapting and producing digital dental implant bridges and denture prostheses based on CBCT and intraoral scan data, which is used to realize the automated design, adaptation and production of dental implant bridges and denture prostheses. The specific steps of the method include: Step 1, data acquisition: Use CBCT scanning to obtain the three-dimensional anatomical data of the patient's jawbone and implants, and use an oral scanner to obtain the intraoral scanning data of the patient to generate a three-dimensional STL file; Step 2, data fusion: Precisely match the three-dimensional STL file and the three-dimensional anatomical data to generate a complete three-dimensional oral digital model; Step 3, automatic design: Use deep learning-based automatic modeling technology to analyze the patient's occlusion relationship, adjacent tooth morphology, and implant angle, and automatically design dental implants and denture prostheses to obtain a three-dimensional model; Step 4, automatic adaptation: Simulate the dynamic contact between the dental bridge and the patient's actual oral environment to verify the occlusal balance; Step 5, production output: Use high-precision numerical control processing equipment or 3D printing equipment to process dental implants and denture prostheses according to the designed three-dimensional model; Step 6, quality inspection: Use computer vision technology or three-dimensional scanning technology to detect the consistency between the size and shape of the production finished product and the three-dimensional model; Step 7, clinical trial fitting and adjustment: Perform digital trial fitting through augmented reality (AR) technology for dentists to pre-simulate the installation to reduce the clinical adjustment time.

[0008] Preferably, the specific steps of the said Step 1, data acquisition include: Step 1.1, Use an oral scanner to obtain the intraoral scanning data of the patient to generate a high-precision three-dimensional STL file; Step 1.2, Automatically repair the scanning blind area using algorithms such as image reconstruction, interpolation, and edge detection to improve the integrity of the STL data; Step 1.3, Use CBCT scanning to obtain the three-dimensional anatomical data of the patient's jawbone and implants; Step 1.4, Use a compatible algorithm for CT scanner data format, and adopt an image segmentation-based algorithm to extract the key anatomical features around the implant to obtain bone density, alveolar bone morphology, nerve distribution, and implant position data, which are used as the basic data in the automatic design process of dental implants and denture prostheses.

[0009] Preferably, the specific steps of the said Step 2, data fusion include: Step 2.1, Adopt an AI algorithm to automatically identify the angle, depth, and jawbone condition of the implant, and establish a digital coordinate system for the implant restoration area; Step 2.2, Process the medical images obtained by CBCT scanning and extract the semantic features in the images; Step 2.3, Use one-dimensional convolution to process the point cloud data in the STL file, and extract global features through global max pooling to represent the abstract features of the point cloud; Step 2.4: Accurately register the CBCT medical image and the point cloud data in the STL file through fiducial point matching to generate a complete three-dimensional oral digital model.

[0010] Preferably, the data fusion registration technology for the CBCT medical image and the point cloud data in the STL file in step 2.4 includes: Rigid registration based on feature points, using anatomical feature points for point cloud matching to improve the initial registration accuracy; Non-rigid registration based on deep learning, using a deformation network to perform elastic alignment on the intraoral scan STL data and the CBCT data to improve the fitness; Through fiducial point matching and one-dimensional convolutional point cloud processing to improve the point cloud matching accuracy between the CBCT medical image and the STL data, ensuring the consistency of the three-dimensional model.

[0011] Preferably, the specific steps of the third step, automatic design, include: Step 3.1: Define a generator Generator and a discriminator Discriminator based on the generative adversarial network GAN architecture. Step 3.2: Build an automatic generation engine for the dental bridge and denture models based on a large-scale deep learning network of training data, and use the generative adversarial network to optimize the morphological details of the prosthesis to ensure the verisimilitude and biomechanical matching. Step 3.3: Use the Open3D library to realize the visualization of the fused three-dimensional point cloud data and achieve the rendering display of the generation result.

[0012] Preferably, in step 3.2, based on a large-scale deep learning network of training data, the specific steps of building an automatic generation engine for the dental bridge and denture models and using the generative adversarial network to optimize the morphological details of the prosthesis to ensure the verisimilitude and biomechanical matching include: Step 3.2.1: Automatically identify the occlusal relationship and tooth morphology, and generate the best occlusal surface and bridge support structure. Step 3.2.2: Simulate the oral dynamic contact process and optimize the design to avoid occlusal interference. Step 3.2.3: Through the alternating adversarial training mechanism of alternately optimizing the generator Generator and the discriminator Discriminator, realize multi-modal data fusion.

[0013] Preferably, the specific steps of the fourth step, automatic adaptation, include: Step 4.1: Accurately adapt the three-dimensional model of the prosthesis obtained by automatic design to the implant abutment. Step 4.2: Conduct geometric matching analysis on the three-dimensional model of the prosthesis and the implant abutment; perform three-dimensional morphological comparison and surface deviation analysis based on the ICP algorithm. Step 4.3, perform mechanical simulation tests on the three-dimensional model of the prosthesis; simulate the dynamic contact between the three-dimensional model of the prosthesis and the patient's oral environment to verify the occlusal relationship between the prosthesis and the corresponding teeth; Step 4.4, adjust the parameters of the three-dimensional model of the prosthesis according to the occlusal relationship between the prosthesis and the corresponding teeth to improve the adaptability of the three-dimensional model of the prosthesis.

[0014] And, a digital implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data, used to implement the digital implant bridge and denture automatic adaptation and production method based on CBCT and intraoral scan data as described above. The system includes: A data acquisition module, used to obtain the three-dimensional anatomical data of the patient's jawbone and implants through CBCT scanning, and obtain the intraoral scan data of the patient through an oral scanner, and realize the optimization of intraoral scan data and the standardization of CBCT data; A data fusion module, used to accurately match the intraoral scan data and the three-dimensional anatomical data through a multi-level registration algorithm based on feature points, combined with the precise positioning points of the implants and the dental anatomical feature points, to generate a complete three-dimensional oral digital model; An automatic design module, used to automatically generate a prosthesis model by analyzing the patient's occlusal relationship, adjacent tooth morphology, and implant angle; An adaptation optimization module, used for the adaptation modeling of the bridge and the implant abutment. Through the parameter optimization of the angle, depth, and clearance, and the simulated stress analysis of the prosthesis, ensure the stability of the prosthesis; A digital production module, used for the production and processing of the prosthesis, using a CNC machining tool for high-precision cutting, or using 3D printing technology to realize the printing and forming of various materials; A quality inspection module, used to detect the size and shape of the processed finished product through a high-precision three-dimensional scanner. At the same time, detect whether there are surface defects in the product through a computer vision system; A clinical adjustment module, used to simulate the installation effect of the prosthesis through AR technology, simulate the patient's occlusal process, and intuitively display the occlusal balance state, facilitating the doctor and the patient to jointly confirm the treatment plan.

[0015] Preferably, the data acquisition module includes: An intraoral scan data acquisition module, used to obtain intraoral scan data through an oral scanner, generate a three-dimensional STL file, and optimize the scan boundary through an integrated real-time edge detection technology to improve the integrity of the STL data; CBCT data acquisition module, which is used to obtain the three-dimensional anatomical data of the patient's jawbone and implants through CBCT scanning, and extract the key anatomical features around the implants through an image segmentation algorithm, including bone density, alveolar bone morphology, and nerve distribution.

[0016] Preferably, the data fusion module includes: Fusion data generation module, which is used to efficiently register the oral scan data and three-dimensional anatomical data to generate a complete three-dimensional digital oral model; Fusion data verification module, which is used to automatically detect the data fusion accuracy of the oral scan data and three-dimensional anatomical data through a three-dimensional registration accuracy evaluation tool.

[0017] Preferably, the adaptation and optimization module includes: Adaptation algorithm module, which is used for precise modeling of the connection between the bridge and the implant abutment, and ensures the stability and long-term applicability of the prosthesis through parameter optimization; Mechanical simulation module, which is used to simulate the mechanical distribution of the prosthesis during chewing through finite element analysis (FEA), and adjust the design to improve the fatigue resistance of the bridge.

[0018] Preferably, the digital production module includes: CNC machining control module, which is used for high-precision cutting of metal materials or ceramic materials, and reduces machining errors by automatically calibrating the equipment parameters of the CNC machining machine tool; 3D printing control module, which is used for 3D printing of resin materials, and improves the 3D printing accuracy and production efficiency through an intelligent slicing algorithm.

[0019] In the above-mentioned digital dental implant bridge and denture automatic adaptation and production method and system based on CBCT and oral scan data, the CBCT data and oral scan data are fused to generate a three-dimensional digital oral model; the occlusal relationship, adjacent tooth morphology, and implant angle of the patient are analyzed using an automatic modeling technology based on deep learning, and the three-dimensional model of the dental implant bridge and denture is automatically designed to obtain the prosthesis. Computer vision technology or three-dimensional scanning technology is used to detect the consistency between the size and shape of the production finished product and the three-dimensional model; and digital try-on is carried out through augmented reality (AR) technology to quickly adjust the various parameters of the product. The technical solution of the present invention uses an algorithm and process for implant restoration design by combining CBCT and oral scan data, realizes an AI optimization method for automated design and adaptation, and also uses a modeling tool for automated occlusal balance simulation and mechanical optimization to realize a model-free full-process technology from data acquisition to product production. The technical solution of the present invention is simple, easy to implement, low in cost, and convenient for popularization. Description of the Drawings

[0020] Figure 1It is a flowchart of the control method of the digital dental implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data in an embodiment of the present invention.

[0021] Figure 2 It is a block diagram of the composition structure of the digital dental implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data in an embodiment of the present invention. Detailed implementation manners

[0022] In this embodiment, taking the digital dental implant bridge and denture automatic adaptation and production method and system based on CBCT and intraoral scan data as an example, the present invention will be described in detail below in conjunction with specific embodiments and drawings.

[0023] Please refer to Figure 1 , which shows a digital dental implant bridge and denture automatic adaptation and production method provided by an embodiment of the present invention, used to realize the automatic design, adaptation and production of dental implant bridges and dentures. The specific steps of the method include: Step 1, data acquisition; obtaining the three-dimensional anatomical data of the patient's jawbone and implants by CBCT scanning, and using an intraoral scanner to obtain the intraoral scan data of the patient to generate a three-dimensional STL file.

[0024] The specific steps of data acquisition include: Step 1.1, using an intraoral scanner to obtain the intraoral scan data of the patient to generate a high-precision three-dimensional STL file.

[0025] Step 1.2, automatically repairing the scanning blind area by using a point cloud completion algorithm based on a neural network, a reconstruction algorithm based on surface interpolation or a morphological expansion completion algorithm to improve the integrity of the STL data.

[0026] Specifically, during the intraoral scanning process, data loss may occur due to scanning angle, light occlusion or scanning device limitations, such as, subgingival areas, adjacent tooth occlusion areas, etc. For this reason, the following intelligent algorithms can be used to automatically repair the scanning blind area and improve the integrity of the STL data.

[0027] 1) Point cloud completion algorithm based on neural network (Point Completion Network, PCN): using a deep learning network (such as variational autoencoder VAE) to generate point cloud data of the missing area and infer the blind area structure from the existing data.

[0028] 2) Reconstruction algorithm based on surface interpolation: using NURBS (Non-Uniform Rational B-Spline) or Poisson reconstruction method to fit a reasonable missing part through the surrounding known point cloud data.

[0029] 3) Morphology-based extension and completion method: Using the morphological features of neighboring points, perform completion through regularized interpolation or nearest neighbor point prediction.

[0030] The above intelligent algorithms have their own advantages and disadvantages, and are used in combination in practical applications to give full play to their advantages and improve the integrity and quality of the scanning data of CBCT and oral scanners.

[0031] Specifically, the oral scanner integrates real-time edge detection technology to optimize the scanning boundary and improve the efficiency of oral scanning. Among them, the automatically repaired scanning blind areas include the subgingival area, etc., to obtain complete data for oral scanning.

[0032] Step 1.3, Use CBCT scanning to obtain the three-dimensional anatomical data of the patient's jawbone and implant.

[0033] Step 1.4, Use a compatible algorithm for the CT scanner data format, and adopt an image segmentation-based algorithm to extract the key anatomical features around the implant to obtain bone density, alveolar bone morphology and nerve distribution, and implant position data, which are used as basic data in the automatic design process of dental bridges and denture prostheses.

[0034] Specifically, use a compatible algorithm for the CT scanner data format (such as DICOM), which supports the processing of data collected by various different devices, and can obtain the three-dimensional anatomical data of the patient's jawbone and the relevant data of the implant.

[0035] Step two, data fusion; precisely match the three-dimensional STL file and the three-dimensional anatomical data to generate a complete three-dimensional oral digital model.

[0036] The specific steps of data fusion include: Step 2.1, Use an AI algorithm to automatically identify the angle, depth and jawbone condition of the implant, and establish a digital coordinate system for the implant restoration area.

[0037] Step 2.2, Process the medical images obtained by CBCT scanning and extract the semantic features in the images.

[0038] Step 2.3, Use one-dimensional convolution to process the point cloud data in the STL file, and extract global features through global max pooling to represent the abstract features of the point cloud.

[0039] Step 2.4, Through landmark point matching, precisely register the CBCT medical images and the point cloud data in the STL file to generate a complete three-dimensional oral digital model.

[0040] Specifically, the data fusion and registration technology for the CBCT medical images and the point cloud data in the STL file in the above steps includes: Feature - point - based Rigid Registration: Using anatomical feature points (such as the root apex, tooth edge, implant vertex) for point - cloud matching to improve the initial registration accuracy.

[0041] Deep - learning - based Non - rigid Registration: Using deformation networks (such as PointNet++, CPD algorithm) to perform elastic alignment on the oral - scan STL data and CBCT data to improve the adaptability.

[0042] Through fiducial point matching and one - dimensional convolutional point - cloud processing: Improving the point - cloud matching accuracy of CBCT medical images and STL data to ensure the consistency of the 3D model.

[0043] Step 3, Automatic Design; Using deep - learning - based automatic modeling technology to analyze the patient's occlusion relationship, adjacent tooth morphology, and implant angle, and automatically design dental bridges and denture prostheses to obtain a 3D model.

[0044] The specific steps of automatic design include: Step 3.1, Based on the Generative Adversarial Network (GAN) architecture, define the generator Generator and discriminator Discriminator.

[0045] Specifically, during the automatic design of dental bridges and denture prostheses, the Generative Adversarial Network (GAN) is used to optimize the morphology of the restorations to ensure verisimilitude and biomechanical matching: Use the generator (Generator) to create a high - precision 3D model of the restoration; Use the discriminator (Discriminator) to distinguish between real denture models and generated denture models to improve the morphological rationality; Combine large - scale oral data to train the GAN network so that the automatically generated restorations conform to the personalized occlusion relationship and anatomical characteristics.

[0046] Step 3.2, Based on a large - scale deep - learning network of training data, construct an automatic generation engine for dental bridge and denture models, and use the Generative Adversarial Network to optimize the morphological details of the restorations to ensure verisimilitude and biomechanical matching.

[0047] The specific steps include: Step 3.2.1, Automatically identify the occlusion relationship and tooth morphology, and generate the optimal occlusal surface and bridge support structure.

[0048] Step 3.2.2, Simulate the oral dynamic contact process, such as chewing movement, etc., and optimize the design to avoid occlusal interference.

[0049] Step 3.2.3, Through the alternating adversarial training mechanism of alternately optimizing the generator Generator and discriminator Discriminator, achieve multi - modal data fusion.

[0050] Step 3.3: Use the Open3D library to visualize the fused three-dimensional point cloud data and render and display the generated results.

[0051] Step Four: Automatic adaptation; simulate the dynamic contact between the dental bridge and the patient's actual oral environment to verify the occlusal balance.

[0052] The specific steps of automatic adaptation include: Step 4.1: Precisely adapt the three-dimensional model of the prosthesis obtained by automatic design to the implant abutment.

[0053] Step 4.2: Conduct geometric matching analysis on the three-dimensional model of the prosthesis and the implant abutment; perform three-dimensional shape comparison and surface deviation analysis based on the ICP algorithm; Three-dimensional shape comparison based on the ICP (Iterative Closest Point) algorithm: Use the ICP algorithm to calculate the registration error between the bridge and the abutment to ensure that the fitting error is below the micron level.

[0054] Surface deviation analysis: Use computer vision or optical scanning techniques, such as white light scanning or laser scanning, to measure the tiny gap between the bridge and the abutment and visualize it through an error heat map.

[0055] Optimize the parameters of the angle, depth, and gap of the three-dimensional model of the prosthesis so that the connection part can be seamlessly connected and stably installed.

[0056] Step 4.3: Conduct mechanical simulation tests on the three-dimensional model of the prosthesis; simulate the dynamic contact between the three-dimensional model of the prosthesis and the patient's oral environment to verify the occlusal relationship between the prosthesis and the corresponding teeth.

[0057] Specifically, based on the finite element analysis (FEA) algorithm, simulate the stress situation during chewing and analyze whether the bridge can evenly distribute the occlusal force to prevent fracture caused by stress concentration.

[0058] Dynamic occlusion simulation: Use dynamic mechanical simulation software, such as Ansys and Abaqus, to test the stability of the bridge under different angles and occlusal forces.

[0059] Step 4.4: Adjust the parameters of the three-dimensional model of the prosthesis according to the occlusal relationship between the prosthesis and the corresponding teeth to improve the adaptability of the three-dimensional model of the prosthesis.

[0060] Step Five: Production output; process dental implants and denture prostheses according to the designed three-dimensional model through high-precision numerical control processing equipment or 3D printing equipment.

[0061] Specifically, mechanical load tests and microstructure inspections are required for the produced dental implants and denture prostheses.

[0062] Mechanical load test. Through tensile test or pressure test, verify whether the bridge meets the ISO 14801 implant connection durability standard.

[0063] Microstructure detection. Use an electron microscope (SEM) to analyze the contact interface between the bridge and the abutment, and observe whether there are microcracks or surface defects.

[0064] Step 6, quality inspection; Through computer vision technology or 3D scanning technology, detect the consistency between the size and shape of the produced finished product and the 3D model.

[0065] Specifically, if the quality inspection is unqualified, that is, the size and shape of the produced finished product are inconsistent with the 3D model, it is necessary to analyze the reasons for the defective products and re - conduct production and processing.

[0066] Step 7, clinical trial fitting and adjustment; Through augmented reality (AR) technology for digital trial fitting, for dentists to pre - simulate the installation to reduce the clinical adjustment time.

[0067] Specifically, in the technical solution of the digital implant bridge and denture automatic adaptation and production method based on CBCT and intraoral scan data of the present invention, the core technical points include: 1), Data fusion technology: Propose a multi - source data fusion method combining CBCT and intraoral scan data, and use implant positioning points and anatomical feature points for high - precision registration to ensure the anatomical integrity and spatial consistency of the model.

[0068] 2), AI - driven automatic design and optimization: Based on deep learning models and biomechanical analysis, automatically generate highly adaptable implant restorations, shortening the design cycle and improving the adaptation accuracy.

[0069] 3), Model - free production process: Combine virtual verification and digital processing, completely eliminating the traditional physical model printing link, and realizing fast and accurate production.

[0070] And, a digital implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data, used to implement the digital implant bridge and denture automatic adaptation and production method based on CBCT and intraoral scan data as described above. The system includes: Data acquisition module 10, used to obtain the three - dimensional anatomical data of the patient's jawbone and implants through CBCT scanning, and obtain the intraoral scan data of the patient through an oral scanner, and realize the optimization of intraoral scan data and the standardization of CBCT data.

[0071] The data fusion module 20 is used to accurately match the intraoral scan data and the three-dimensional anatomical data by means of a multi-level registration algorithm based on feature points, combining the precise positioning points of the implant and the dental anatomical feature points, and generating a complete three-dimensional digital oral model.

[0072] The automatic design module 30 is used to automatically generate a prosthesis model by analyzing the patient's occlusion relationship, the morphology of adjacent teeth and the implant angle.

[0073] The adaptation and optimization module 40 is used for the adaptation modeling of the bridge and the implant abutment. By optimizing the parameters of the angle, depth and clearance, and the simulated stress analysis of the prosthesis, the stability of the prosthesis is ensured.

[0074] Specifically, the adaptation and optimization module 40 can achieve the dynamic optimization and adaptation of complex multi-abutment connecting bridges.

[0075] The digital production module 50 is used for the production and processing of the prosthesis. High-precision cutting is carried out by using a CNC machine tool, or 3D printing technology is used to realize the printing and forming of various materials.

[0076] The quality inspection module 60 is used to detect the size and shape of the processed finished product by means of a high-precision three-dimensional scanner. At the same time, whether there are surface defects on the product is detected through a computer vision system.

[0077] Specifically, the surface defects of the product include cracks, surface roughness, etc.

[0078] Specifically, the detection results of the quality inspection module 60 are transmitted back to the system in real time to optimize the subsequent production design.

[0079] The clinical adjustment module 70 is used to simulate the installation effect of the prosthesis through AR technology, simulate the patient's occlusion process, intuitively display the occlusion balance state, and facilitate the doctor and the patient to jointly confirm the treatment plan.

[0080] Specifically, the clinical adjustment module 70 also provides a digital fine-tuning tool, allowing the doctor to quickly optimize the design parameters based on the patient's feedback.

[0081] Preferably, the data acquisition module 10 includes: The intraoral scan data acquisition module is used to obtain the intraoral scan data through an oral scanner, generate a three-dimensional STL file, and optimize the scan boundary by integrating real-time edge detection technology to improve the integrity of the STL data; The CBCT data acquisition module is used to obtain the three-dimensional anatomical data of the patient's jawbone and implant through CBCT scanning, and extract the key anatomical features around the implant through an image segmentation algorithm, including bone density, alveolar bone morphology and nerve distribution.

[0082] Preferably, the data fusion module 20 includes: A fused data generation module, configured to efficiently register oral scan data and three-dimensional anatomical data to generate a complete three-dimensional digital oral model; A fused data verification module, configured to automatically detect the data fusion accuracy of the oral scan data and the three-dimensional anatomical data through a three-dimensional registration accuracy evaluation tool.

[0083] Preferably, the adaptation and optimization module 40 includes: An adaptation algorithm module, configured to perform precise modeling of the connection between the bridge and the implant abutment, and ensure the stability and long-term applicability of the prosthesis through parameter optimization; A mechanical simulation module, configured to simulate the mechanical distribution of the prosthesis during chewing through finite element analysis (FEA), and adjust the design to improve the fatigue resistance of the bridge.

[0084] Preferably, the digital production module 50 includes: A CNC machining control module, configured to perform high-precision cutting of metal materials or ceramic materials, and reduce machining errors by automatically calibrating the equipment parameters of the CNC machining machine; A 3D printing control module, configured to perform 3D printing of resin materials, and improve the 3D printing accuracy and production efficiency through an intelligent slicing algorithm.

[0085] Specifically, the data fusion module is used to fuse oral scan point cloud data (STL) and CBCT image data (DICOM), and its data fusion process is as follows: import torch import torch.nn as nn import torch.nn.functional as F from torchvision.models.segmentation import fcn_resnet50 import open3d as o3d # Feature Extraction for CBCT (2D Medical Images) class CBCTFeatureExtractor(nn.Module): def __init__(self): super(CBCTFeatureExtractor, self).__init__() self.model = fcn_resnet50(pretrained=False, num_classes=2) def forward(self, x): return self.model(x)["out"] # Feature Extraction for STL (Point Cloud Data) class PointCloudFeatureExtractor(nn.Module): def __init__(self): super(PointCloudFeatureExtractor, self).__init__() self.conv1 = nn.Conv1d(3, 64, 1) self.conv2 = nn.Conv1d(64, 128, 1) self.conv3 = nn.Conv1d(128, 256, 1) self.fc = nn.Linear(256, 128) def forward(self, x): x = F.relu(self.conv1(x)) x = F.relu(self.conv2(x)) x = F.relu(self.conv3(x)) x = torch.max(x, 2, keepdim=False)[0]# Global max pooling x = self.fc(x) return x # Fusion Network class DataFusionNetwork(nn.Module): def __init__(self): super(DataFusionNetwork, self).__init__() self.cbct_extractor = CBCTFeatureExtractor() self.pointcloud_extractor = PointCloudFeatureExtractor() self.fc = nn.Linear(256, 128) def forward(self, cbct, pointcloud): cbct_features = self.cbct_extractor(cbct) pointcloud_features = self.pointcloud_extractor(pointcloud) fused_features = torch.cat([cbct_features, pointcloud_features], dim=1) fused_output = self.fc(fused_features) return fused_output Specifically, the prosthesis automatic design module adopts an automatic design method based on the GAN adversarial network architecture, and its automatic design process is as follows: # Generator Network class Generator(nn.Module): def __init__(self): super(Generator, self).__init__() self.fc1 = nn.Linear(128, 512) self.fc2 = nn.Linear(512, 1024) self.fc3 = nn.Linear(1024, 2048) self.fc4 = nn.Linear(2048, 3 * 1024)# Output 3D point cloud def forward(self, x): x = F.relu(self.fc1(x)) x = F.relu(self.fc2(x)) x = F.relu(self.fc3(x)) x = torch.tanh(self.fc4(x))# Normalize output return x.view(-1, 1024, 3)# Point cloud with 1024 points # Discriminator Network class Discriminator(nn.Module): def __init__(self): super(Discriminator, self).__init__() self.fc1 = nn.Linear(3 * 1024, 1024) self.fc2 = nn.Linear(1024, 512) self.fc3 = nn.Linear(512, 128) self.fc4 = nn.Linear(128, 1) def forward(self, x): x = x.view(-1, 3 * 1024)# Flatten point cloud x = F.leaky_relu(self.fc1(x), 0.2) x = F.leaky_relu(self.fc2(x), 0.2) x = F.leaky_relu(self.fc3(x), 0.2) return torch.sigmoid(self.fc4(x)) Among them, the training process is as follows: # Hyperparameters lr = 0.0002 batch_size = 16 epochs = 100 # Initialize models fusion_net = DataFusionNetwork() generator = Generator() discriminator = Discriminator() # Optimizers opt_g = torch.optim.Adam(generator.parameters(), lr=lr, betas=(0.5,0.999)) opt_d = torch.optim.Adam(discriminator.parameters(), lr=lr, betas=(0.5, 0.999)) # Loss functions adversarial_loss = nn.BCELoss() # Training Loop for epoch in range(epochs): for cbct_data, pointcloud_data, real_data in dataloader: # Fuse data fused_features = fusion_net(cbct_data, pointcloud_data) # Train Discriminator real_labels = torch.ones(batch_size, 1) fake_labels = torch.zeros(batch_size, 1) real_outputs = discriminator(real_data) fake_data = generator(fused_features).detach() fake_outputs = discriminator(fake_data) d_loss_real = adversarial_loss(real_outputs, real_labels) d_loss_fake = adversarial_loss(fake_outputs, fake_labels) d_loss = d_loss_real + d_loss_fake opt_d.zero_grad() d_loss.backward() opt_d.step() # Train Generator fake_data = generator(fused_features) fake_outputs = discriminator(fake_data) g_loss = adversarial_loss(fake_outputs, real_labels) opt_g.zero_grad() g_loss.backward() opt_g.step() print(f"Epoch [{epoch+1} / {epochs}] - D Loss: {d_loss.item():.4f}, GLoss: {g_loss.item():.4f}") Specifically, the visualization process of the 3D model result of the prosthesis automatically designed is as follows: def visualize_point_cloud(point_cloud): pc = o3d.geometry.PointCloud() pc.points = o3d.utility.Vector3dVector(point_cloud) o3d.visualization.draw_geometries([pc]) # Example usage test_input = torch.randn(1, 128)# Random test input generated_point_cloud = generator(test_input).detach().numpy()[0] visualize_point_cloud(generated_point_cloud) In the above-mentioned method and system for automatic adaptation and production of digital implant bridges and dentures based on CBCT and oral scan data, CBCT data and oral scan data are fused to generate a three-dimensional oral digital model; the patient's occlusal relationship, adjacent tooth morphology and implant angle are analyzed using automatic modeling technology based on deep learning, and the implant bridge and denture are automatically designed to obtain a three-dimensional model of the restoration. Computer vision technology or three-dimensional scanning technology is used to detect the consistency of the size and shape of the finished product with the three-dimensional model; and digital trial wearing is performed through augmented reality AR technology to achieve rapid adjustment of various parameters of the product. The technical solution of the present invention uses algorithms and processes for implant restoration design combined with CBCT and oral scan data to realize an AI optimization method for automated design and adaptation, and also uses modeling tools for automated occlusal balance simulation and mechanical optimization to realize a full-process technology from data acquisition to product production without modeling. The technical solution of the present invention is simple, easy to implement, low-cost, and easy to promote.

[0086] It should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A digital implant bridge and denture automatic adaptation and production method based on CBCT and intraoral scan data, which is used to realize the automatic design, adaptation and production of implant bridges and dentures. It is characterized in that The specific steps of the method include: Step 1, data acquisition; obtaining three-dimensional anatomical data of the patient's jawbone and implants by CBCT scanning, and obtaining intraoral scanning data of the patient using an oral scanner to generate a three-dimensional STL file; Step 2, data fusion; precisely matching the three-dimensional STL file and the three-dimensional anatomical data to generate a complete three-dimensional oral digital model; Step 3, automatic design; using an automatic modeling technology based on deep learning to analyze the patient's occlusion relationship, adjacent tooth morphology, and implant angle, and automatically designing dental implant bridges and denture prostheses to obtain a three-dimensional model of the restoration; Step 4, automatic adaptation; simulating the dynamic contact between the dental bridge and the patient's actual oral environment to verify the occlusal balance; Step 5, production output; machining dental implant bridges and denture prostheses according to the designed three-dimensional model through high-precision numerical control machining equipment or 3D printing equipment; Step 6, quality inspection; detecting the consistency between the size and shape of the production finished product and the three-dimensional model through computer vision technology or three-dimensional scanning technology; Step 7, clinical trial fitting and adjustment; performing digital trial fitting through augmented reality (AR) technology for dental doctors to pre-simulate the installation to reduce the clinical adjustment time.

2. The automatic adaptation and production method of digital implant bridges and denture false teeth based on CBCT and oral scan data according to claim 1, characterized in that, The specific steps of the said Step 1, data acquisition, include: Step 1.1, obtaining intraoral scanning data of the patient using an oral scanner to generate a high-precision three-dimensional STL file; Step 1.2, automatically repairing the scanning blind area using algorithms such as image reconstruction, interpolation, and edge detection to improve the integrity of the STL data; Step 1.3, obtaining three-dimensional anatomical data of the patient's jawbone and implants by CBCT scanning; Step 1.4, using a compatible algorithm for CT scanner data format, and adopting an image segmentation algorithm to extract key anatomical features around the implant to obtain bone density, alveolar bone morphology, nerve distribution, and implant position data, which are used as basic data in the automatic design process of dental bridges and denture prostheses.

3. The automatic adaptation and production method of digital implant bridges and denture false teeth based on CBCT and intraoral scan data according to claim 1, characterized in that, The specific steps of the said Step 2, data fusion, include: Step 2.1, using an AI algorithm to automatically identify the angle, depth, and jawbone condition of the implant, and establishing a digital coordinate system for the implant restoration area; Step 2.2, processing the medical images obtained by CBCT scanning to extract semantic features in the images; Step 2.3, using one-dimensional convolution to process the point cloud data in the STL file, and extracting global features through global max pooling to represent the abstract features of the point cloud; Step 2.4, precisely registering the CBCT medical images and the point cloud data in the STL file through fiducial point matching to generate a complete three-dimensional oral digital model.

4. The automatic adaptation and production method of digital implant bridges and denture false teeth based on CBCT and intraoral scan data according to claim 3, characterized in that, The data fusion registration technology for the CBCT medical images and the point cloud data in the STL file in the said Step 2.4 includes: Rigid registration based on feature points, using anatomical feature points for point cloud matching to improve the initial registration accuracy; Non-rigid registration based on deep learning, using a deformation network to perform elastic alignment on the intraoral scan STL data and CBCT data to improve the adaptability; Through fiducial point matching and one-dimensional convolution point cloud processing, to improve the point cloud matching accuracy of CBCT medical images and STL data, and ensure the consistency of the three-dimensional model.

5. The automatic adaptation and production method of digital dental implants, bridges and dentures based on CBCT and intraoral scan data according to claim 1, characterized in that, The specific steps of the third step, i.e., automatic design, include: Step 3.1: Based on the Generative Adversarial Network (GAN) architecture, define the Generator and the Discriminator. Step 3.2: Based on a large-scale deep learning network of training data, construct an automatic generation engine for dental bridge and denture models, and use the Generative Adversarial Network to optimize the morphological details of the prosthesis to ensure verisimilitude and biomechanical matching. Step 3.3: Use the Open3D library to visualize the fused three-dimensional point cloud data and realize the rendering display of the generation results.

6. The automatic adaptation and production method of digital dental implants, bridges and dentures based on CBCT and intraoral scan data according to claim 5, characterized in that, For the said Step 3.2, the specific steps of constructing an automatic generation engine for dental bridge and denture models based on a large-scale deep learning network of training data, and using the Generative Adversarial Network to optimize the morphological details of the prosthesis to ensure verisimilitude and biomechanical matching include: Step 3.2.1: Automatically identify the occlusion relationship and tooth morphology, and generate the optimal occlusal surface and bridge support structure. Step 3.2.2: Simulate the dynamic contact process of the oral cavity, and optimize the design to avoid occlusal interference. Step 3.2.3: Through the alternating adversarial training mechanism of alternately optimizing the Generator and the Discriminator, realize multi-modal data fusion.

7. The automatic adaptation and production method of digital dental implants, bridges and dentures based on CBCT and intraoral scan data according to claim 1, characterized in that, The specific steps of the fourth step, i.e., automatic adaptation, include: Step 4.1: Perform precise adaptation design on the three-dimensional model of the prosthesis obtained by automatic design and the implant abutment. Step 4.2: Conduct geometric matching analysis on the three-dimensional model of the prosthesis and the implant abutment; perform three-dimensional shape comparison and surface deviation analysis based on the ICP algorithm. Step 4.3: Conduct mechanical simulation tests on the three-dimensional model of the prosthesis; simulate the dynamic contact between the three-dimensional model of the prosthesis and the patient's oral environment to verify the occlusion relationship between the prosthesis and the corresponding teeth. Step 4.4: Adjust the parameters of the three-dimensional model of the prosthesis according to the occlusion relationship between the prosthesis and the corresponding teeth to improve the adaptability of the three-dimensional model of the prosthesis.

8. A digital implant bridge and denture automatic adaptation and production system based on CBCT and oral scan data, which is used to implement the digital implant bridge and denture automatic adaptation and production method based on CBCT and oral scan data as described in any one of claims 1-7, and is characterized in that, It includes: A data acquisition module, which is used to obtain the three-dimensional anatomical data of the patient's jawbone and implants through CBCT scanning, and obtain the intraoral scanning data of the patient through an oral scanner, and realize the optimization of the oral scan data and the standardization of the CBCT data. A data fusion module, which is used to precisely match the oral scan data and the three-dimensional anatomical data through a multi-level registration algorithm based on feature points, combined with the precise positioning points of the implants and the dental anatomical feature points, to generate a complete three-dimensional digital oral model. An automatic design module, which is used to automatically generate a prosthesis model by analyzing the patient's occlusion relationship, adjacent tooth morphology, and implant angle. An adaptation optimization module, which is used for the adaptation modeling of the bridge and the implant abutment, and ensures the stability of the prosthesis through parameter optimization of the angle, depth, and clearance, as well as the simulated stress analysis of the prosthesis. A digital production module, which is used for the production and processing of the prosthesis, and uses a CNC machining tool for high-precision cutting, or uses 3D printing technology to realize the printing and forming of various materials. A quality inspection module, which is used to detect the size and morphology of the processed finished product through a high-precision three-dimensional scanner, and at the same time, detect whether there are surface defects in the product through a computer vision system. A clinical adjustment module, which is used to simulate the installation effect of a prosthesis through AR technology, simulate the patient's occlusion process, visually display the occlusion balance state, and facilitate the doctor and the patient to jointly confirm the treatment plan.

9. The digital implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data according to claim 8, characterized in that, The data acquisition module includes: An intraoral scan data acquisition module, which is used to obtain intraoral scan data through an oral scanner, generate a three-dimensional STL file, and optimize the scan boundary through integrated real-time edge detection technology to improve the integrity of the STL data; A CBCT data acquisition module, which is used to obtain three-dimensional anatomical data of the patient's jawbone and implants through CBCT scanning, and extract key anatomical features around the implants through an image segmentation algorithm, including bone density, alveolar bone morphology, and nerve distribution; The data fusion module includes: A fused data generation module, which is used to efficiently register the intraoral scan data and the three-dimensional anatomical data to generate a complete three-dimensional oral digital model; A fused data verification module, which is used to automatically detect the data fusion accuracy of the intraoral scan data and the three-dimensional anatomical data through a three-dimensional registration accuracy evaluation tool.

10. The digital implant bridge and denture automatic adaptation and production system based on CBCT and intraoral scan data according to claim 8, characterized in that, The adaptation and optimization module includes: An adaptation algorithm module, which is used for precise modeling of the connection between the bridge and the implant abutment, and ensures the stability and long-term applicability of the prosthesis through parameter optimization; A mechanical simulation module, which is used to simulate the mechanical distribution of the prosthesis during chewing through finite element analysis (FEA), and adjust the design to improve the fatigue resistance of the bridge; The digital production module includes: A CNC machining control module, which is used for high-precision cutting of metal materials or ceramic materials, and reduces machining errors by automatically calibrating the equipment parameters of the CNC machining machine tool; A 3D printing control module, which is used for 3D printing of resin materials, and improves the 3D printing accuracy and production efficiency through an intelligent slicing algorithm.

Images